def main(**job_inputs):
input_vcfs = [
dx_utils.download_and_gunzip_file(f, skip_decompress=True)
for f in job_inputs['input_vcfs']
]
input_ref = dx_utils.download_and_gunzip_file(job_inputs['ref_fasta'])
# create index files for inputs
dx_utils.run_cmd(['samtools', 'faidx', input_ref])
map(dx_utils.run_cmd, ['tabix {0}'.format(vcf) for vcf in input_vcfs])
with open(VCF_FOFN, 'w') as fh:
fh.write('\n'.join(input_vcfs))
# get the bcftools version and help doc
cmd = ['bcftools', '--help']
dx_utils.run_cmd(cmd)
output_prefix = job_inputs.get('output_prefix', '')
output_bcf = output_prefix + 'concat' + '.bcf'
# concatenate the bcf/vcf files
concat_cmd = ['bcftools', 'concat', '-f', VCF_FOFN]
view_cmd = ['bcftools', 'view', '-Ou', '-e\'type="ref"\'']
norm_cmd = [
'bcftools', 'norm', '-Ob', '-f', input_ref, '-o', output_bcf,
'--threads={0}'.format(multiprocessing.cpu_count())
]
# run the commands
dx_utils.run_pipe(concat_cmd, view_cmd, norm_cmd)
# index the concatenated bcf file
dx_utils.run_cmd(['bcftools', 'index', output_bcf])
# call consensus
output_fasta = output_prefix + 'consensus.fasta'
consensus_filter = 'QUAL>1 && (GT="AA" || GT="Aa")'
consensus_cmd = [
'bcftools', 'consensus', '-i', consensus_filter, '-Hla', '-f',
input_ref, output_bcf
]
dx_utils.run_pipe(consensus_cmd, outputFile=output_fasta)
# save the changes to vcf
output_vcf = output_prefix + 'changes.vcf.gz'
vcf_cmd = [
'bcftools', 'view', '-i', consensus_filter, '-Oz',
'--threads={0}'.format(multiprocessing.cpu_count()), output_bcf
]
dx_utils.run_pipe(vcf_cmd, outputFile=output_vcf)
consensus_link = dx_utils.gzip_and_upload(output_fasta)
print(consensus_link)
output = {}
output['consensus_fasta'] = consensus_link
output['consensus_vcf'] = dxpy.dxlink(dxpy.upload_local_file(output_vcf))
return output
def main(input_assembly,
hic_alignments,
restriction_enzyme_bases,
filter_alignments,
input_assembly_graph=None):
# make sure we can run salsa
dx_utils.run_cmd("python /opt/SALSA/run_pipeline.py -h")
input_assembly = dx_utils.download_and_gunzip_file(input_assembly)
alignment_prefix = input_assembly.split(".fasta")[0]
# process inputs and convert to bed
first_file = True
for bam_file in hic_alignments:
fn = dxpy.describe(bam_file['$dnanexus_link'])['name']
cmd = 'dx cat {0}'.format(bam_file['$dnanexus_link'])
prefix, suffix = os.path.splitext(fn)
if suffix == '.gz':
cmd += '| gunzip '
fn = prefix
cmd += '| bedtools bamtobed -i stdin'
if first_file:
cmd += ' > {0}.bed'.format(alignment_prefix)
first_file = False
else:
cmd += ' >> {0}.bed'.format(alignment_prefix)
dx_utils.run_cmd(cmd)
# index the ref
cmd = 'samtools faidx {0} '.format(input_assembly)
dx_utils.run_cmd(cmd)
# if we were asked to filter by contig names, make a bed file and subset the input bed
if filter_alignments == True:
f = open('%s.fai' % (input_assembly))
o = open("%s.contigs.bed" % (alignment_prefix), 'w')
for line in f:
line = line.strip().split()
o.write("%s\t1\t%s\n" % (line[0], line[1]))
f.close()
o.close()
cmd = 'bedtools intersect -wa -a {0}.bed -b {0}.contigs.bed > {0}.filtered.bed'.format(
alignment_prefix)
dx_utils.run_cmd(cmd)
else:
cmd = 'ln -s {0}.bed {0}.filtered.bed'.format(alignment_prefix)
dx_utils.run_cmd(cmd)
# now sort the bed file
cmd = "sort -T . -k4 {0}.filtered.bed > {0}.sorted.bed".format(
alignment_prefix)
dx_utils.run_cmd(cmd)
cmd = 'python /opt/SALSA/run_pipeline.py -a {0} -b {1}.sorted.bed -l {0}.fai -o {2} -e {3} -m yes -p yes '
cmd = cmd.format(input_assembly, alignment_prefix, './',
','.join(restriction_enzyme_bases))
if input_assembly_graph is not None:
cmd = "%s -g input_assembly_graph" % (cmd)
dx_utils.run_cmd(cmd)
output = {}
# final scaffold
final_fasta = glob.glob('scaffold*FINAL.fasta')[0]
output['final_scaffold_fasta'] = dx_utils.gzip_and_upload(final_fasta)
# final agp
final_agp = glob.glob('scaffold*FINAL.agp')[0]
output['final_scaffold_agp'] = dx_utils.gzip_and_upload(final_agp)
# alignment_iteration_1 bed
alignment_iteration_1 = glob.glob('alignment_iteration_1.bed')[0]
output['alignment_iteration_1'] = dx_utils.gzip_and_upload(
alignment_iteration_1)
# scaffold_length_iteration_1
scaffold_length_iteration_1 = glob.glob('scaffold_length_iteration_1')[0]
output['scaffold_length_iteration_1'] = dx_utils.gzip_and_upload(
scaffold_length_iteration_1)
# all others
files = glob.glob('scaffold*fasta')
files.extend(glob.glob('scaffold*agp'))
if final_fasta in files:
files.remove(final_fasta)
if final_agp in files:
files.remove(final_agp)
print files
output['scaffold'] = dx_utils.tar_files_and_upload(files, alignment_prefix)
return output
def main(**job_inputs):
bionano_cmap_1_link = job_inputs['bng_enzyme1']
bionano_cmap_2_link = job_inputs['bng_enzyme2']
ngs_fasta_link = job_inputs['ngs_fasta_or_cmap']
args_xml_link = job_inputs.get('args_xml')
# Download all the inputs
bionano_cmap_1_filename = os.path.join(
'/home/dnanexus', download_and_gunzip_file(bionano_cmap_1_link))
bionano_cmap_2_filename = os.path.join(
'/home/dnanexus', download_and_gunzip_file(bionano_cmap_2_link))
ngs_fasta_filename = os.path.join('/home/dnanexus',
download_and_gunzip_file(ngs_fasta_link))
if args_xml_link:
args_xml_filename = download_and_gunzip_file(args_xml_link)
else:
args_xml_filename = os.path.join(HYBRID_DIR, 'TGH',
'hybridScaffold_two_enzymes.xml')
output_dir = "hybrid_scaffold_output"
run_cmd('mkdir {0}'.format(output_dir))
results_tar = output_dir + '_results.tar'
cmd = "Rscript {dir}/runTGH.R --help".format(dir=HYBRID_DIR)
run_cmd(cmd)
scaffold_cmd = (
"Rscript {dir}/runTGH.R -N {ngs_fasta} "
"-b1 {bng1} -b2 {bng2} -O {outdir} -R {refaligner} -t {results} ".
format(dir=HYBRID_DIR,
ngs_fasta=ngs_fasta_filename,
bng1=bionano_cmap_1_filename,
bng2=bionano_cmap_2_filename,
outdir=output_dir,
refaligner=os.path.join(TOOLS_DIR, 'RefAligner'),
results=results_tar))
scaffold_cmd += '-e1 {enzyme1} -e2 {enzyme2} '.format(
enzyme1=job_inputs['enzyme1_name'], enzyme2=job_inputs['enzyme2_name'])
if job_inputs.get("cuts1_file") and job_inputs.get("cuts2_file"):
cuts1_file = download_and_gunzip_file(job_inputs["cuts1_file"])
cuts2_file = download_and_gunzip_file(job_inputs["cuts2_file"])
scaffold_cmd += '-m1 {cuts1} -m2 {cuts2} '.format(cuts1=cuts1_file,
cuts2=cuts2_file)
scaffold_cmd += ' {args_xml}'.format(args_xml=args_xml_filename)
run_cmd(scaffold_cmd)
scaffold_final = glob.glob(
os.path.join(output_dir, 'TGH_M1', 'AGPExport', '*HYBRID*'))
if not scaffold_final:
print("ERROR: No hybrid scaffolds produced.")
hybrid_scaffold_log = os.path.join(output_dir, 'TGH.log')
run_cmd('tail -n 50 {0}'.format(hybrid_scaffold_log))
scaffold_final_ncbi = glob.glob(
os.path.join(output_dir, 'hybrid_scaffolds*',
'*_HYBRID_SCAFFOLD_NCBI.fasta'))
unscaffolded_final = glob.glob(
os.path.join(output_dir, 'hybrid_scaffolds*',
'*_HYBRID_SCAFFOLD_NOT_SCAFFOLDED.fasta'))
output = {
"scaffold_fasta": [
dxpy.dxlink(dxpy.upload_local_file(f)) for f in scaffold_final
if f.endswith(".fasta")
],
"scaffold_output":
[dxpy.dxlink(dxpy.upload_local_file(f)) for f in scaffold_final],
"ncbi_scaffold_final":
dx_utils.gzip_and_upload(scaffold_final_ncbi[0]),
"unscaffolded_final":
dx_utils.gzip_and_upload(unscaffolded_final[0])
}
tar_name = "hybrid_scaffold_output.tar.gz"
tar_cmd = "tar czvf {tar_name} {outdir}".format(tar_name=tar_name,
outdir=output_dir)
run_cmd(tar_cmd)
output_id = dxpy.upload_local_file(tar_name)
output["scaffold_targz"] = dxpy.dxlink(output_id)
return output
def main(**job_inputs):
bionano_cmap_link = job_inputs['refinefinal_merged_cmap']
ngs_fasta_link = job_inputs['ngs_fasta_or_cmap']
args_xml_link = job_inputs.get('args_xml')
# Download all the inputs
bionano_cmap_filename = download_and_gunzip_file(bionano_cmap_link)
ngs_fasta_filename = download_and_gunzip_file(ngs_fasta_link)
if args_xml_link:
args_xml_filename = download_and_gunzip_file(args_xml_link)
else:
args_xml_filename = os.path.join(HYBRID_DIR, 'hybridScaffold_config.xml')
output_dir = "hybrid_scaffold_output"
scaffold_cmd = (
"perl {dir}/hybridScaffold.pl -n {ngs_fasta} -b {cmap} "
"-o {outdir} -c {args_xml} "
.format(dir=HYBRID_DIR, ngs_fasta=ngs_fasta_filename, cmap=bionano_cmap_filename,
args_xml=args_xml_filename, outdir=output_dir))
scaffold_cmd += '-r {refaligner} '.format(refaligner=os.path.join(TOOLS_DIR, 'RefAligner'))
if "conflict_resolution_file" in job_inputs:
conflict_resolution_file = download_and_gunzip_file(job_inputs["conflict_resolution_file"])
scaffold_cmd += '-M {cr_file} '.format(conflict_resolution_file)
else:
scaffold_cmd += '-B {b_level} -N {n_level} '.format(
b_level=job_inputs["b_conflict_filter"], n_level=job_inputs["n_conflict_filter"])
if job_inputs["generate_molecules"] is True:
scaffold_cmd += '-x '
scaffold_cmd += '-p {0}'.format(SCRIPTS_DIR)
try:
molecules_bnx_file = download_and_gunzip_file(job_inputs["molecules_bnx_file"])
scaffold_cmd += '-m {0} '.format(molecules_bnx_file)
except KeyError:
raise dxpy.AppError("Molecules BNX file required for Align Molecules flag (-x)")
try:
optargs_xml = download_and_gunzip_file(job_inputs["optargs_xml"])
scaffold_cmd += '-q {0} '.format(optargs_xml)
except KeyError:
raise dxpy.AppError("OptArgs XML file required for Align Molecules flag (-x)")
if job_inputs["generate_chimeric"] is True:
scaffold_cmd += '-y '
if molecules_bnx_file:
scaffold_cmd += '-m {0} '.format(molecules_bnx_file)
else:
try:
molecules_bnx_file = download_and_gunzip_file(job_inputs["molecules_bnx_file"])
scaffold_cmd += '-m {0} '.format(molecules_bnx_file)
except KeyError:
raise dxpy.AppError("Molecules BNX file required for Generate Molecules flag")
if "err_files" in job_inputs:
err_files = [download_and_gunzip_file(err_file) for err_file in job_inputs["err_files"]]
err_cmd = ' '.join(['-e {0}'.format(err) for err in err_files])
scaffold_cmd += err_cmd
run_cmd(scaffold_cmd)
run_cmd('tree {0}'.format(output_dir))
scaffold_final_ncbi = glob.glob(
os.path.join(output_dir, 'hybrid_scaffolds*', '*_HYBRID_SCAFFOLD_NCBI.fasta'))
unscaffolded_final = glob.glob(
os.path.join(output_dir, 'hybrid_scaffolds*', '*_HYBRID_SCAFFOLD_NOT_SCAFFOLDED.fasta'))
scaffold_final = glob.glob(
os.path.join(output_dir, 'hybrid_scaffolds*', '*_HYBRID_SCAFFOLD.fasta'))
scaffold_final.extend(glob.glob(
os.path.join(output_dir, 'hybrid_scaffolds*', '*_HYBRID_SCAFFOLD.cmap')))
scaffold_final.extend(glob.glob(
os.path.join(output_dir, 'hybrid_scaffolds*', '*_HYBRID_SCAFFOLD.agp')))
scaffold_output = glob.glob(
os.path.join(output_dir, 'hybrid_scaffolds*', '*_HYBRID_SCAFFOLD*'))
cut_and_conflict = glob.glob(os.path.join(output_dir, 'hybrid_scaffolds*', 'conflicts*.txt'))
cut_and_conflict.extend(glob.glob(os.path.join(output_dir, 'hybrid_scaffolds*', '*_annotations.bed')))
output = {"scaffold_final": [dx_utils.gzip_and_upload(f) for f in scaffold_final],
"scaffold_output": [dx_utils.gzip_and_upload(f) for f in scaffold_output if f not in scaffold_final],
"cut_and_conflict": [dxpy.dxlink(dxpy.upload_local_file(f)) for f in cut_and_conflict],
"ncbi_scaffold_final": dx_utils.gzip_and_upload(scaffold_final_ncbi[0]),
"unscaffolded_final": dx_utils.gzip_and_upload(unscaffolded_final[0])}
tar_name = "hybrid_scaffold_output.tar.gz"
tar_cmd = "tar czvf {tar_name} {outdir}".format(
tar_name=tar_name,
outdir=output_dir)
dx_utils.run_cmd(tar_cmd)
output_id = dxpy.upload_local_file(tar_name)
output["scaffold_targz"] = dxpy.dxlink(output_id)
return output
def main(**job_inputs):
bionano_cmap_link = job_inputs['refinefinal_merged_cmap']
ngs_fasta_link = job_inputs['ngs_fasta_or_cmap']
args_xml_link = job_inputs.get('args_xml')
# Download all the inputs
bionano_cmap_filename = dx_utils.download_and_gunzip_file(bionano_cmap_link)
ngs_fasta_filename = dx_utils.download_and_gunzip_file(ngs_fasta_link)
if args_xml_link:
args_xml_filename = dx_utils.download_and_gunzip_file(args_xml_link)
else:
args_xml_filename = os.path.join(HYBRID_DIR, 'hybridScaffold_config.xml')
output_dir = "hybrid_scaffold_output"
scaffold_cmd = ["perl", os.path.join(HYBRID_DIR, "hybridScaffold.pl"), "-n", ngs_fasta_filename,
"-b", bionano_cmap_filename, "-o", output_dir, "-c", args_xml_filename,
"-r", os.path.join(TOOLS_DIR, 'RefAligner')]
if "conflict_resolution_file" in job_inputs:
conflict_resolution_file = dx_utils.download_and_gunzip_file(job_inputs["conflict_resolution_file"])
scaffold_cmd += ["-M", conflict_resolution_file]
else:
scaffold_cmd += ["-B", str(job_inputs["b_conflict_filter"]), "-N", str(job_inputs["n_conflict_filter"])]
molecules_bnx_file = None
if job_inputs["generate_molecules"] is True:
scaffold_cmd += ["-x", "-p", SCRIPTS_DIR]
try:
molecules_bnx_file = dx_utils.download_and_gunzip_file(job_inputs["molecules_bnx_file"])
scaffold_cmd += ["-m", molecules_bnx_file]
except KeyError:
raise dxpy.AppError("Molecules BNX file required for Align Molecules flag (-x)")
try:
optargs_xml = dx_utils.download_and_gunzip_file(job_inputs["optargs_xml"])
scaffold_cmd += ["-q", optargs_xml]
except KeyError:
raise dxpy.AppError("OptArgs XML file required for Align Molecules flag (-x)")
if job_inputs["generate_chimeric"] is True:
scaffold_cmd += ["-y"]
if molecules_bnx_file:
scaffold_cmd += ["-m", molecules_bnx_file]
else:
try:
molecules_bnx_file = dx_utils.download_and_gunzip_file(job_inputs["molecules_bnx_file"])
scaffold_cmd += ["-m", molecules_bnx_file]
except KeyError:
raise dxpy.AppError("Molecules BNX file required for Generate Molecules flag")
if "err_files" in job_inputs:
err_files = [dx_utils.download_and_gunzip_file(err_file) for err_file in job_inputs["err_files"]]
for err in err_files:
scaffold_cmd += ["-e", err]
dx_utils.run_cmd(scaffold_cmd)
dx_utils.run_cmd(["tree", output_dir])
scaffold_final_ncbi = glob.glob(
os.path.join(output_dir, 'hybrid_scaffolds', '*_HYBRID_SCAFFOLD_NCBI.fasta'))[0]
unscaffolded_final = glob.glob(
os.path.join(output_dir, 'hybrid_scaffolds', '*_HYBRID_SCAFFOLD_NOT_SCAFFOLDED.fasta'))[0]
scaffold_final = glob.glob(
os.path.join(output_dir, 'hybrid_scaffolds', '*_HYBRID_SCAFFOLD.fasta'))
scaffold_final.extend(glob.glob(
os.path.join(output_dir, 'hybrid_scaffolds', '*_HYBRID_SCAFFOLD.cmap')))
scaffold_final.extend(glob.glob(
os.path.join(output_dir, 'hybrid_scaffolds', '*_HYBRID_SCAFFOLD.agp')))
scaffold_output = glob.glob(os.path.join(output_dir, 'hybrid_scaffolds', '*_HYBRID_SCAFFOLD.xmap'))
scaffold_output.extend(glob.glob(os.path.join(output_dir, 'hybrid_scaffolds', '*_HYBRID_SCAFFOLD_q.cmap')))
scaffold_output.extend(glob.glob(os.path.join(output_dir, 'hybrid_scaffolds', '*_HYBRID_SCAFFOLD_r.cmap')))
scaffold_output = [f for f in scaffold_output if f not in scaffold_final]
cut_and_conflict = glob.glob(os.path.join(output_dir, 'hybrid_scaffolds*', 'conflicts*.txt'))
cut_and_conflict.extend(glob.glob(os.path.join(output_dir, 'hybrid_scaffolds*', '*_annotations.bed')))
# make sure output files don't have colons
dx_utils.run_cmd(["sed", "-i.bak", "s/:/_/g", scaffold_final_ncbi])
dx_utils.run_cmd(["sed", "-i.bak", "s/:/_/g", unscaffolded_final])
# upload outputs
output = {"scaffold_final": [dx_utils.gzip_and_upload(f) for f in scaffold_final],
"scaffold_output": [dx_utils.gzip_and_upload(f) for f in scaffold_output],
"cut_and_conflict": [dxpy.dxlink(dxpy.upload_local_file(f)) for f in cut_and_conflict],
"ncbi_scaffold_final": dx_utils.gzip_and_upload(scaffold_final_ncbi),
"unscaffolded_final": dx_utils.gzip_and_upload(unscaffolded_final)}
tar_name = "hybrid_scaffold_output.tar.gz"
tar_cmd = "tar czvf {tar_name} {outdir}".format(
tar_name=tar_name,
outdir=output_dir)
dx_utils.run_cmd(tar_cmd)
output_id = dxpy.upload_local_file(tar_name)
output["scaffold_targz"] = dxpy.dxlink(output_id)
return output
def main(**job_inputs):
bionano_cmap_1_link = job_inputs['bng_enzyme1']
bionano_cmap_2_link = job_inputs['bng_enzyme2']
ngs_fasta_link = job_inputs['ngs_fasta_or_cmap']
args_xml_link = job_inputs.get('args_xml')
# Download all the inputs
bionano_cmap_1_filename = os.path.join(
'/home/dnanexus', dx_utils.download_and_gunzip_file(bionano_cmap_1_link))
bionano_cmap_2_filename = os.path.join(
'/home/dnanexus', dx_utils.download_and_gunzip_file(bionano_cmap_2_link))
ngs_fasta_filename = os.path.join('/home/dnanexus', dx_utils.download_and_gunzip_file(ngs_fasta_link))
if args_xml_link:
args_xml_filename = dx_utils.download_and_gunzip_file(args_xml_link)
else:
args_xml_filename = os.path.join(HYBRID_DIR, 'TGH', 'hybridScaffold_two_enzymes.xml')
output_dir = "hybrid_scaffold_output"
dx_utils.run_cmd(['mkdir', output_dir])
results_tar = output_dir + '_results.tar'
cmd = ["Rscript", os.path.join(HYBRID_DIR, "runTGH.R"), "--help"]
dx_utils.run_cmd(cmd)
scaffold_cmd = ["Rscript", os.path.join(HYBRID_DIR, "runTGH.R"), "-N", ngs_fasta_filename,
"-b1", bionano_cmap_1_filename, "-b2", bionano_cmap_2_filename, "-O", output_dir,
"-R", os.path.join(TOOLS_DIR, 'RefAligner'), "-t", results_tar,
"-e1", job_inputs['enzyme1_name'], "-e2", job_inputs['enzyme2_name']]
if job_inputs.get("cuts1_file") and job_inputs.get("cuts2_file"):
cuts1_file = dx_utils.download_and_gunzip_file(job_inputs["cuts1_file"])
cuts2_file = dx_utils.download_and_gunzip_file(job_inputs["cuts2_file"])
scaffold_cmd += ["-m1", cuts1_file, "-m2", cuts2_file]
scaffold_cmd += [args_xml_filename]
dx_utils.run_cmd(scaffold_cmd)
# try locating the outputs
final_dirs = ["TGH_M2", "TGH_M1", "two_enzyme_hybrid_scaffold_M2", "two_enzyme_hybrid_scaffold_M1"]
for possible_loc in final_dirs:
scaffold_final = glob.glob(os.path.join(output_dir, possible_loc, 'AGPExport', '*HYBRID_Export.fasta'))
if scaffold_final:
scaffold_final_ncbi = glob.glob(
os.path.join(output_dir, possible_loc, 'AGPExport', '*HYBRID_Export_NCBI.fasta'))[0]
unscaffolded_final = glob.glob(
os.path.join(output_dir, possible_loc, 'AGPExport', '*HYBRID_Export_NOT_SCAFFOLDED.fasta'))[0]
scaffold_output = glob.glob(os.path.join(output_dir, possible_loc, '*_HYBRID_Export.agp'))
scaffold_output.extend(glob.glob(os.path.join(output_dir, possible_loc, '*_HYBRID_Export.xmap')))
scaffold_output.extend(glob.glob(os.path.join(output_dir, possible_loc, '*_HYBRID_Export_q.cmap')))
scaffold_output.extend(glob.glob(os.path.join(output_dir, possible_loc, '*_HYBRID_Export_r.cmap')))
scaffold_output = [f for f in scaffold_output if f not in scaffold_final]
break
# if still not found, something went wrong
if not scaffold_final:
hybrid_scaffold_log = os.path.join(output_dir, 'TGH.log')
dx_utils.run_cmd(["tail", "-n", "50", hybrid_scaffold_log])
raise dxpy.AppError("ERROR: No hybrid scaffolds produced.")
# make sure output files don't have colons
dx_utils.run_cmd(["sed", "-i.bak", "s/:/_/g", scaffold_final_ncbi])
dx_utils.run_cmd(["sed", "-i.bak", "s/:/_/g", unscaffolded_final])
output = {
"scaffold_fasta": [dxpy.dxlink(dxpy.upload_local_file(f)) for f in scaffold_final if f.endswith(".fasta")],
"scaffold_output": [dxpy.dxlink(dxpy.upload_local_file(f)) for f in scaffold_output],
"ncbi_scaffold_final": dx_utils.gzip_and_upload(scaffold_final_ncbi),
"unscaffolded_final": dx_utils.gzip_and_upload(unscaffolded_final)
}
tar_name = "hybrid_scaffold_output.tar.gz"
tar_cmd = "tar czvf {tar_name} {outdir}".format(
tar_name=tar_name,
outdir=output_dir)
dx_utils.run_cmd(tar_cmd)
output_id = dxpy.upload_local_file(tar_name)
output["scaffold_targz"] = dxpy.dxlink(output_id)
return output